| dc.contributor.author |
Ghahramani, Zoubin |
en_US |
| dc.contributor.author |
Jordan, Michael I. |
en_US |
| dc.date.accessioned |
2004-10-20T20:49:37Z |
|
| dc.date.available |
2004-10-20T20:49:37Z |
|
| dc.date.issued |
1995-01-24 |
en_US |
| dc.identifier.other |
AIM-1509 |
en_US |
| dc.identifier.other |
CBCL-108 |
en_US |
| dc.identifier.uri |
http://hdl.handle.net/1721.1/7202 |
|
| dc.description.abstract |
Real-world learning tasks often involve high-dimensional data sets with complex patterns of missing features. In this paper we review the problem of learning from incomplete data from two statistical perspectives---the likelihood-based and the Bayesian. The goal is two-fold: to place current neural network approaches to missing data within a statistical framework, and to describe a set of algorithms, derived from the likelihood-based framework, that handle clustering, classification, and function approximation from incomplete data in a principled and efficient manner. These algorithms are based on mixture modeling and make two distinct appeals to the Expectation-Maximization (EM) principle (Dempster, Laird, and Rubin 1977)---both for the estimation of mixture components and for coping with the missing data. |
en_US |
| dc.description.provenance |
Made available in DSpace on 2004-10-20T20:49:37Z (GMT). No. of bitstreams: 2
AIM-1509.ps: 388268 bytes, checksum: 16da10c310f72f441702d15b47a79750 (MD5)
AIM-1509.pdf: 515095 bytes, checksum: a068c9e2a95a4179cd9886fe2a62ef65 (MD5)
Previous issue date: 1995-01-24 |
en |
| dc.format.extent |
11 p. |
en_US |
| dc.format.extent |
388268 bytes |
|
| dc.format.extent |
515095 bytes |
|
| dc.format.mimetype |
application/postscript |
|
| dc.format.mimetype |
application/pdf |
|
| dc.language.iso |
en_US |
|
| dc.relation.ispartofseries |
AIM-1509 |
en_US |
| dc.relation.ispartofseries |
CBCL-108 |
en_US |
| dc.subject |
AI |
en_US |
| dc.subject |
MIT |
en_US |
| dc.subject |
Artificial Intelligence |
en_US |
| dc.subject |
missing data |
en_US |
| dc.subject |
mixture models |
en_US |
| dc.subject |
statistical learning |
en_US |
| dc.subject |
EM algorithm |
en_US |
| dc.subject |
maximum likelihood |
en_US |
| dc.subject |
neural networks |
en_US |
| dc.title |
Learning from Incomplete Data |
en_US |
